human capital measurement: country experiences and international...
TRANSCRIPT
1
Human capital measurement:
country experiences and international initiatives
Gang Liu
Statistics Norway
Presented at the Third World KLEMS Conference
Tokyo, Japan, May 19-20, 2014
1
Presentation outline
1. Concept and definition
2. Implications for measurement
3. Country experiences
4. International activities
5. Main issues and challenges
6. Concluding remarks
2
1. Concept and Definition
• Roots can be found in the history of economic thought: Petty (1690),
Smith (1776), Farr (1853), and Engel (1883).
• Recognition regained since 1960s: Schultz (1961), Becker (1964) and
Mincer (1974).
• The OECD definition: ‘the knowledge, skills, competencies and
attributes embodied in individuals that facilitate the creation of personal,
social and economic well-being’ (OECD, 2001)
• An all-embracing definition that has obtained wide acceptance.
3
1. Concept and Definition (cont.)
4
Box 1. Human capital: a sketch of its formation, composition and benefits generated
Informal learning
On-the-job training
Health care
Parenting
Education
Human capital investment
(both lifelong and life-wide)
Human capital embodied
in individuals
Benefits due to human capital investment
Knowledge
Skills
Competencies
Attributes
Economic
Non-economic (Personal)
(Personal)
Non-economic
(Social)
(Social)
Market activities
Non-market activities
Health
Subjective well-being
…………
Informed citizens
Willingness to cooperate
…………
Migration
…………
2. Implications for measurement
• Stepwise approach: starting from those aspects of either lower
conceptual challenges or greater data availability.
• Distinguishing health capital from human capital.
• Focusing on formal education (as the main form of human capital
investment); and on the economic returns to the individual (as the main
benefits due to human capital investment), even if the broader OECD
definition is accepted as a useful reference point.
5
2. Implications for measurement (Cont.)
6
Box 2. Inputs, outputs and outcomes of education sector
Inputs Outputs
Outcomes
Number of students/ schooling
years by level of education
(Visible)
with human capital
embodied/ accumulated
(Invisible)
Direct outcomes: Test scores
(e.g. results of pencil and paper tests)
Indirect outcomes: Economic benefits
Non-economic benefits (Personal) Non-economic benefits (Social)
Labor
Capital
Intermediate
consumptions
(Including
both market
and non-
market inputs)
Environmental factors (e.g. innate abilities, cultural, social, and economic backgrounds, as well as
political, legal and institutional arrangements)
Both direct and indirect outcomes can be used for quality adjustment for outputs
Box 3. Classification of measuring methodologies
Human capital
measurement
Indicators-based
approach
Monetary
measures
Quantitative indicators
Qualitative indicators
Cost-based approach
Income-based approach
Residual approach
3. Country experiences
3.1 Results of the UNECE CES questionnaire on measuring human capital
• Purpose
• Concept
• Methodology
• Data sources
• Status of the estimates
7
3. Country experiences (Cont.)
3.2 Representative studies using the indicators-based approach
• Single indicators : adult literacy rates (e.g. Azariadis and Drazen, 1990;
Romer, 1990), school enrolment ratios (e.g. Barro, 1991; Mankiw et al.,
1992), average years of schooling (e.g.Temple, 1999; Krueger and
Lindahl, 2001).
• Dashboard type indicators (e.g. Education at a Glance; Ederer et al.,
2007, 2011)
• Advantages: simple, less data-demanding
• Disadvantages: lack of common metric, not really accounting
8
3. Country experiences (Cont.)
3.3 Representative studies using the cost-based approach
• Advantages: data availability, PIM
• Disadvantages: cost of production not necessarily equal market value,
investment-consumption dichotomy, choice of depreciation
• Most well-known studies: Kendrick (1976), Eisner (1985)
• Recent national studies: Germany (Ewerhart, 2001, 2003), the
Netherlands (Rooijen-Horsten et al., 2007, 2008), Finland (Kokkinen,
2008, 2010)
9
3. Country experiences (Cont.)
3.4 Representative studies using the income-based approach
• Advantages: theoretically sound, practically feasible, possible to be
incorporated into the SNA in the future
• Disadvantages: no perfect labor market, choice of key parameters
• Most well-known studies: lifetime income approach by Jorgenson and
Fraumeni (1989, 1992a, 1992b)
10
3. Country experiences (Cont.) 3.4 Representative studies using the income-based approach (Cont.)
11
Examples of
national
studies
Country Motivation Time range Main data sources Population
covered
Market/
Nonmarket
activities
Jorgenson and
Fraumeni
(1989, 1992a,
1992b)
United
States
New systems of
national accounts,
Output of education
sector
1948-1984,
1947-1987
Rich data based on
decades of research
Age 0-75 Both
Ahlroth, et al
(1997)
Sweden Output of education
sector
1967, 1973,
1980, 1990
Level of living
surveys
Age 0-75 Both
Ervik, et al
(2003)
Norway Output of higher
education sector
1995 Register data Age 20-64 Market only
Wei (2004,
2008)
Australia Incorporating human
capital into the SNA
(Stock/Flow)
1981-2001 Census data Age 18 (25)-65,
labor
force/whole
population
Market only
Le, et al (2006) New
Zealand
Measuring human
capital (Stock)
1981-2001 Census data Age 18-64 Market only
Gundimeda, et
al (2006)
India Accounting for human
capital formation
1993-2001 Surveys of
employment and
unemployment,
Census of population
Age 15-60 Market only
Gu and Wong
(2008)
Canada Human capital
contribution to national
wealth account
1970-2007 Census /labour force
survey
Age 15-74 Market only
Liu and
Greaker (2009)
Norway Measuring human
capital (Stock)
2006 Register data Age 15(16)-
67(74), labor
force/ whole
population
Market only
Christian
(2010)
United
States
Measuring human
capital
(Stock/Investment)
1994-2006
Rich data Age 0-80 Both
Coremberg
(2010)
Argentina Measuring human
capital (Stock)/Output
of education sector
1997, 2001,
2004
Household
permanent survey
Age 15-65 Market only
Li, et al. (2010) China Measuring human
capital (Stock)
1985-2007 Household
survey/Health and
nutrition survey
Urban/rural,
Age 0-60 (55 for
female)
Market only
Jones and
Chiripanhura
(2010)
United
Kingdom
Measuring human
capital (Stock)
2001-2009 Labor force survey Age 16-64 Market only
Istat (2013) Italy Measuring human
capital (Stock)
2008 Various surveys Age 15-64 Both
4. International initiatives
• Indicators-based approach: Barro and Lee (1993, 1996, 2001, 2010,
2013); OECD (Education at a Glance, PISA, PIAAC); UN (works on
constructing sustainable development indices, HDI)
• Residual approach: World Bank (2006, 2011)
• UN Inclusive Wealth Report (UN-IHDP, UNEP, 2012)
• Lifetime income approach: OECD human capital project (Liu, 2011)
• Joint work by the World Bank and the OECD (Hamilton and Liu, 2014)
12
4. International initiatives (Cont.)
13
0
100
200
300
400
500
600
700
800
900
ROU POL ITA ISR NZL ESP NLD DNK AUS FRA KOR JPN CAN NOR GBR USA
Human capital per capita (US$ in thousands) GDP per capita (US$ in hundreds)
Graph 1: Human capital per capita in 2006 (in thousands US dollars)
Note: Estimates for Australia refer to 2001 and for Denmark to 2002. Source: OECD human capital project.
4. International initiatives (Cont.)
14
Graph 2. Stock of human capital relative to GDP and to the stock of produced capital, 2006
Panel a. Stock of human capital to GDP Panel b. Stock of human capital to produced capital
0,0
4,0
8,0
12,0
16,0
20,0
NLD ITA FRA USA ROU CAN NOR DNK GBR ISR NZL AUS ESP JPN POL KOR
0,0
2,0
4,0
6,0
8,0
ITA NLD FRA DNK USA ESP AUS CAN NZL GBR
Note: Estimates for Australia refer to 2001, those for Denmark to 2002.
Source: OECD human capital project.
4. International initiatives (Cont.)
15
Graph 3. Decomposition of average annual growth of human capital volume per capita due to age, gender and educational attainment (first-order partial volume index, percentages)
-1,0
-0,5
0,0
0,5
1,0
1,5
2,0
AUS (1997-2
001)
CAN (1997-2
006)
FRA (1998-2007)
ISR (2002-2
007)
ITA (1998-2
006)
JPN (2
002-2007)
KOR (1998-2007)
NZL (1997-2
007)
NOR (1997-2
006)
POL (1999-2006)
ESP (2001-2
006)
GBR (1997-2
007)
USA (1997-2
007)
Age Education Gender HC per capita
Note: For many countries, the contribution from gender is too small to be discernable in the figure.
Source: OECD human capital project.
4. International initiatives (Cont.)
16
Table 2. Country rankings of human capital measured by different approaches
PISA Science
2006
PIIAC Literacy 2011-2
PIAAC Numeracy
2011-2
PIAAC Problem-solving in Tech-rich
Environments 2011-2
Barro-Lee Average
Educational Attainment
2005
Jorgenson-Fraumeni
Human Capital per
Capita 2006
1
World Bank
Intangible Capital 2005
AGES 15 16-65 16-65 16-65 15-64 15-64 All Ages
Australia 3 3 5 4 7 9 6
Canada 4 5 6 5 3 4 7
China 17 18 17
Denmark 10 8 3 3 14 8 2
France 9 11 10 11 7 9
Great Britain 5 7 8 6 16 2 8
India 18 17 18
Israel 8 13 4
Italy 14 13 12 12 14 11
Japan 2 1 1 6 6 6 13
Netherlands 6 2 2 1 9 10 5
New Zealand 1 2 12 3
Norway 12 4 3 2 4 3 1
Poland 11 10 9 10 15 15 14
Romania 10 16 15
South Korea 7 5 7 9 5 5 12
Spain 13 12 13 13 11 10
United States 8 9 11 8 1 1 16
Notes: 1. The J-F figures for Australia and India are for 2001; those for Denmark are for 2002. 2. The ages covered for China include ages 16 through 55 for females and 16 through 59 for males.
The ages covered for India include ages 15 through 60.
5. Main issues and challenges
• Data issues: earnings, enrolments, labor force survey, educational
attainment, mortality rates
• Methodological difficulties: cohort effects, business cycle effects, choice
of key parameters, accounting for divergence between estimates by the
cost-based and the income-based approaches
17
6. Concluding remarks
• The multifaceted nature requires stepwise approach for human capital
measurement, e.g. focusing on formal education and economic returns
to individuals as a point of departure.
• Monetary measures, such as those by the cost-based and the income-
based approaches, and esp. by the lifetime income approach seem to
be most promising to be incorporated into the SNA in the future.
• Continue the work on data compilation and harmonization.
• Modify the methodologies, possibly based on new sources of data.
• Construct experimental satellite accounts.
• Link the estimates of human capital to the standard growth accounting
framework.
• Encourage research on streamlined approach for those countries in
which the needed data is not available.
18